Re: [hybi] Ping/Pong body (was Re: TSV-Directorate review of draft-ietf-hybi-thewebsocketprotocol-07)

[Takeshi Yoshino <[e-mail]>]

On Wed, May 25, 2011 at 12:37, Patrick McManus <[e-mail]> wrote:

> right, without further changes to the protocol. Then we could ship
> websockets and people could start writing applications instead of
> documents. Stability is a worthwhile goal.


I really care the time of shipment, too. I don't want this to be show
stopper, but want to get attention to this point from people who want to use
some extended Ping, unsolicited Pong, etc. to get this thought out enough.


> >  enough, maybe I would use random bytes of 3 or more byte long.
>
> right. The ping-generating-peer doesn't need cooperation from the other
> peer in order to create a unique payload. Interoperabiity does not
>

Yes.


> require that each end use the same algorithm - either the client or the
> server can be the ping generating peer.
>

Sorry, I couldn't get you. A and B don't have to use the same algorithm, but
in fact those algorithms need coordination. A's algorithm must choose X for
its Ping where X != Y which B's algorithm chooses for its unsolicited
Pong. Well, I agree that I'm saying something paranoid, but it's true.

How about this? Semantically a bit smaller change.
"Body of Ping and unsolicited body can be empty. IF Ping or unsolicited Pong
have non-empty body, they must start with 0x00 and 0x01"
This doesn't break interoperability with -07 or add new constraint to body
as long as the user-agent doesn't have any data to be put into body. When
some peer want to disambiguate them, it can safely do it by prefixing body
with 0x00 or 0x01 without any new negotiation, coordination or randomness.

For any foobar,
0x00 foobar != 0x01 foobar
0x00 foobar != ""
0x01 foobar != ""